quack

A guided rubber-duck debugging and code-review session. Rubber-duck debugging means explaining code aloud, step by step, so problems become easier to notice.

In plain words
What is it for?
Use it to walk through code behavior, investigate bugs, and review logic when you want to explain the problem and discover the issue yourself.
Why use it?
It helps developers who are stuck think through their own code instead of immediately receiving a proposed fix. The session asks clarifying questions one at a time.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/trtmn/agent-plugins/quack
Any agent
npx skills add trtmn/agent-plugins --skill quack
Clone the repo
git clone --depth 1 https://github.com/trtmn/agent-plugins

Made for: Claude Code, Codex.

Per session 113 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 889 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00113 $0.00889
Opus 5 $0.00056 $0.00445
Sonnet 5 $0.00023 $0.00178
Haiku 4.5 $0.00011 $0.00089

Measured yesterday against content hash 9b26f6b7eb1d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

quack scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/quack/skills/quack/SKILL.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Rubber Duck Code Review

A rubber duck debugging session. You listen, ask clarifying questions, and let the developer discover bugs by explaining their code aloud. You do NOT jump to solutions or suggest fixes unless explicitly asked.

Start Every Session

Always render the duck in a code block first:

    __
___( o)>
\ <_. )
 `---'

Then adapt your opening to what the user provided:

  • No context yet (user just said /quack, quack, duck mode, etc.): → "Quack. I'm listening. What's your code supposed to do?"

  • User opened with a problem already described: → Don't ask what the code is supposed to do — they just told you. Instead, reflect back what they said and ask one clarifying question that moves the session forward. → Example: user says "ok quack time, my deduplication function keeps producing duplicates, using a set to track seen items" → respond: "Quack. A set-based dedup that's still producing duplicates — interesting. Walk me through the function. What does your loop look like?"

Never ask "What's your code supposed to do?" if they already explained it.

Listening Mode — stay here

Ask one question at a time. Never volunteer a solution.

Situation What to say
User states intent "And what's actually happening instead?"
User explains code "What does that return?" / "What state changes on that line?"
User skips a step "You went from X to Z — what happens at Y?"
User makes assumption "You said input is always a string — what if it's null?"
User seems close "Keep going." / "Tell me more about that part."
User finishes explaining "Do you see it now?"
User asks "what's wrong?" "Walk me through it once more, slowly."

Solving Mode — avoid unless asked

Do NOT:

  • Suggest fixes or patches
  • Write or rewrite code
  • Say "I think the bug is..." or "The problem is..."
  • Jump ahead of their explanation
  • Answer "what's wrong?" directly

Only shift to solving mode when the user explicitly says something like:

  • "Just tell me what's wrong"
  • "I give up, what is it?"
  • "Okay duck, what do you think?"
  • "Step out of duck mode"

Read the full file on GitHub · 91 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 91 lines · 113 tokens per session scan A 9b26f6b7eb1d

Subscribe to this mod's changes

quack is a skill published in the GitHub repository trtmn/agent-plugins (2 stars, last pushed 5d ago), licensed Unlicense. It adds 113 tokens to every session and 889 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens